Document egw8kEMrYbZDR3vrxxzX3nDG

This summary of the Third Annual Report on Car-. ..... . cinogens is published by the National Toxicology Pro gram Public Information Office. The individual entries for the cancer-causing agents contained in this sum mary are reproduced from the Third Annual Report on Carcinogens. Copies of the Third Annual Report on Carcinogens are available for a fee from the National Technical Informa tion Service, Springfield, Virginia 22161. Specify: PB-83-135855. Phone:(703)487-4650. For additional copies of this booklet, contact: NTP Public Information Office, MD B2-04, Box 12233, Research Triangle Park. NC 27709. Phone: (919) 541-3991. Third Annual Report on Carcinogens Summary September, 1983 NTP 82-330 U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES Public Health Service PCB-ARCH-EXT0378747 Contents Page Executive Summary 1 Introduction 5 Substances or groups of substances, and technological or manufacturing processes that are Known to be carcinogenic: 4-Aminobiphenyl 10 Arsenic and certain arsenic compounds . 22 Asbestos 24 Aurarnine manufacture 26 Benzene 28 Benzidine 30 N,N-bis(2-chloroethyl)-2-naphthylamine (chlornaphazine) 34 Bis(chloromethyl)ether (BCME) and technical grade chloromethyl methyl ether (CMME) 35 Chlorambucil 39 Chromium and certain chromiumcompounds 41 Coke oven emissions 42 Cyclophosphamide 45 Diethylstilbestrol (DES) 60 Hematite underground mining 73 Isopropyl alcohol manufacture (strong-acid process) 79 Melphalan 04 Mustard gas 00 2-Naphthylamine 09 Nickel refining 90 Soots, tars, and mineral oils n9 Thorium dioxide 124 Vinyl chloride 132 Substances or groups of substances that may reasonably be anticipated to be carcinogens: 2-Acetylaminofluorene 14 Acrylonitrile 14 Aflatoxins 16 2-Aminoanthraqumone 17 1 -Amino-2-methylanlhraquinone 10 Amitrole 20- o-Anisidine and o-anisidine hydrochloride 20 Aramite 2i Benz(a)anthracene 2/ Benzo(b)fluoranthene 3' Benzo(a)pyrene ' . 5t n~r, >* ^ ' *nin "* I " r*v'\t PCB-ARCH-EXT0378748 Cadmium and certain cadmium compounds Carbon tetrachloride Chloroform p-Cresidine Cupferron Cycasin 2.4-Diaminoanisole sulfate 2.4- Diaminotoluene Dibenz(a.h)acridine pibenz(a,j)acridine Dibenz(a,h)anthracene 7H-Dibenzo(c,g)carbazole Dibenzo(a,h)pyrene Dibenzo(a,i)pyrene 1.2-Dibromo-3-chloropropane (DBCP) 1.2-Dibromoethane (EDB) 3,3'-Dichlorobenzidine 1.2- Dichloroethane (EDC) Diepoxybutane Di(2-ethylhexyl)phthalate 3,3' -Dimethoxybenzidine 4-Dimethylaminoazobenzene 3,3'-Dimethylbenzidine Dimethylcarbamoyl chloride Dimethyl sulfate 1.4- Dioxane Direct Black 38 Direct Blue 6 Ethylene thiourea Formaldehyde Hexachlorobenzene Hydrazine and hydrazine sulfate Hydrazobenzene lndeno(1,2,3-cd)pyrene Iron dextran complex Kepone (chlordecone) Lead acetate and lead phosphate Lindane and other hexachlorocyclohexane isomers 4,4'-Methylenebis(2-chloroaniline) (MOCA) 4,4'-Methylenebis(n.n,-dimethyl)benzenamine Michler's ketone Mirex Nickel and certain nickel compounds Nitrilotriacetic acid 5-Nitro-o-anisidine Page 36 38 40 43 44 45 46 47 48 49 50 51 51 52 53 54 55 57 58 58 61 62 63 64 65 66 67 69 70 71 74 75 77 78 78 80 81 82 84 86 87 88 90 91 92 Nitrofen N-Nitrosodi-n-butylamine N-Nitrosodiethanolamine N-Nitrosodiethylamine N-Nitrosodimethylamine p-Nitrosodiphenylamine N-Nitrosodi-n-propylamine N-Nitroso-n-ethylurea N-Nitroso-n-methylurea N-Nitrosomethylvinylamine N-Nitrosomorpholine N-Nijrosonornicotine N-Nitrosopiperidine N-Nitrosopyrrolidine N-Nitrososarcosine Oxymetholone Phenacetin Phenazopyridine and phenazopyridine hydrochloride Phenytoin and sodium salt of pher ytoin Polybrominated biphenyls (PBBs) Polychlorinated biphenyls (PCBs) Procarbazine and procarbazine hydrochloride p-Propiolactone Reserpine Saccharin Safrole Selenium sulfide Streptozotocin Sulfallate 2,3,7,8-Ietrachlorodibenzo-p-dioxin (TCDD) Thioacetamide Thiourea o-Toluidine and o-toluidine hydrochloride Toxaphene 2,4,6- Trichlorophenol Tris(1-aziridinyl)phosphine sulfide Tris(2,3-dibromopropyl)phosphate Urethane Index "** Acronyms and Abbreviations Units of Measurement Page 94 95 96 97 98 99 100 101 101 102 103 103 104 105 106 107 108 108 109 110 111 113 114 114 115 117 117 120 120 121 122 123 125 126 127 129 129 130 138 134 Agents in italics are new to this Report PCB-ARCH-EXT0378749 Copies of this Report summary may be obtained from the Public Informa tion Office. National Toxicology Pro gram. B2-04, P.O. Box 12233. Research Triangle Park, North Caro lina 27709. Copies of the complete re port are available for a fee from the National Technical Information Ser vice. 5285 Port Royal Road, Spring field. Virginia 22161. Request number PB 83-135855 should be indicated. Phone: (703) 487-4650. 4 Introduction Cancer is the second most common cause of death in the United States. One of every four Americans will suffer from cancer sometime during their lifetime; one in every five will die from cancer. In 1979, more than 400,000 Americans died of cancer.' In addition to the physical and emotional suffer ing caused by cancer, this disease may cost the Nation as much as $30 billion each year in lost production and in come, medical expenses, and re search resources. For these reasons, the American public is concerned about cancer and cancer hazards, es pecially about ways to prevent the oc currence or decrease the incidence of cancers. Contrary to popular belief, cancer is not one but many diseases which may have different causes. Most scientists now consider that about one-third to two-thirds of all cancers are associ-. ated with the environment in which we live and work. In this context, the en vironment is understood as "anything that interacts with humans, including substances eaten, drunk, and smoked, natural and medical radiation, work place exposures, drugs, aspects of sexual behavior, and substances in air, water, and soil."2 Some cultural and behavioral' patterns, apart from de fined habits of smoking and alcohol consumption, may significantly influ ence the development of tumors of the gastrointestinal tract, breast, uterus, and prostate. These aspects of social environment, personal behavior, and habits are sometimes called "life style." Although we rarely know the environmental factors and conditions which are responsible for the devel opment of specific cancers, in some cases we are beginning to have some understanding. Most cancers that are associated with the environment should be avertible, at least potentially. Development of cancer also depends on "host factors." Host factors are at tributes of human organisms associ ated with individual differences in the risk of developing a specified cancer. A host factor may be associated with reduced risk of a cancer, not only with increased risk.3 Examples of host fac tors are immunological and endocrine functions, nutritional status, genetic constitution, age, and sex.4 As Americans have become more and more concerned about the environ ment and cancer, they have repeat edly asked the Federal Government to clarify what is known about cancercausing substances. The public wants to know, and should know, which sub stances are regarded as cancer haz ards in the workplace and in the general environment and how well we are pro tected from exposure to these sub stances. Section 262 of Public Law 95-622 of November 9, 1978* reflects the re quests for this information. Paragraph (4) of this section stipulates that, "The secretary shall publish an annual re port which contains: "(A) a list of all substances (i) which either are known to be carcinogens or may reasonably be anticipated to be carcinogens and (ii) to which a significant number of persons resid- Communiiy Mental Health Centers Act. Amendments. 5 PCB-ARCH-EXT0378751 2. Reasonably anticipated to be car cinogens: A. There is "limited evidence of strong-acid process for the manufac ture of isopropyl alcohol. carcinogenicity" from studies of humans, "which indicates that Human and Animal Studies causal interpretation is credi ble. but that alternative expla nations. such as chance, bias, or confounding, could not ad equately be excluded." or B. There is "sufficient evidence of Some understanding of the-human and animal studies used for identifying carcinogens is necessary to clarify what is meant by the words "known" and "reasonably anticipated." The strong est evidence for a relationship be carcinogenicity" from studies of experimental animals "which indicates that there is an in tween exposure to a suspected substance and cancer in humans comes from epidemiological studies. ]; creased incidence of malignant tumors: (a) in multiple species or strains, or (b) in multiple ex periments (preferably with dif ferent routes of administration or using different dose levels), or (c) to an unusual degree with regard to incidence, site or type of tumor, or age at onset. Ad ditional evidence may be pro vided by data concerning dose response effects, as well as in formation on mutagenicity or chemical structure." Although we have adequate epide- ' miological evidence only for a few car- ' cinogens, we think the patterns seen ( in these studies show how other ani mal carcinogens might affect humans. ;;.. Epidemiological investigations give the * best information on cancer-causing j agents, but such studies of groups of' individuals do have short-comings They may require observations over a1' long time because the latent period : between initial exposure and cancer occurrence may be up to 20-30 years., Intentional, long-term exposures of.; human groups are, of course, unac Most of the substances contained in this Annual Report can be identified chemically. Some carcinogens occur naturally, some are produced when materials are burned, and others are ceptable. As an alternative, scientists make educated predictions about the - effects in humans, based on the ef fects observed in studies with exper imental animals. present in raw materials, or in inter Although test animals may responds mediate and final products of indus differently from humans, experience tries. In some cases only the exposure has shown that almost all known hu associated with a specific technolog man carcinogens also cause cancerj ical or manufacturing process, occu in experimental animals if adequately^ pation, or industry can be associated tested.* This finding is understanda^ with increased incidence of cancer in humans, but the substances respon sible cannot be identified. Examples of such processes are the under ground mining of hematite and the 'The evidence for carcinogenicity in expei,.. mental animals for arsenic is still regarded as inadequate and the evidence for ben2ene is ea! present limited.8 ble since cancer begins in individual cells and all mammalian cells are comparable in structure and function. The basic premise of experimental carcinogenesis research is that sub stances that produce cancer in ro dents and other experimental animals may affect human cells in the same way. Relatively large doses of chemicals are used in animal studies to make up for the small number of experimental an imals exposed, as compared to the size of human populations. Animal tests are costly, take years to complete, and are often difficult to interpret in terms of human carcinogenic risk. They are, nevertheless, the best testing tool we have. No animal test can prove be yond doubt that a substance will or will not cause cancer in humans. Even "sufficient evidence" of carcinogen icity in experimental animals may prove not to be definitive. But if we are to prevent cancer, we cannot wait 20 to 30 years to obtain definitive evidence of harm lo human health. Therefore, substances which show `.'sufficient ev idence" of cancer causation in exper imental animals are treated as if they may reasonably be anticipated to be human carcinogens................ . .. Inclusion of Substances The Third Annual Report contains sub stances, groups of substances, and technological processes which were listed in the Second Annual Report. Where necessary, the information has been updated. The Third Annual Re port presents information on 28 ad ditional substances. 'Each has . been chosen either from substances tested by the NCI Carcinogenesis Testing Program or the NTP; from designa tions of the participating agencies; or from substances evaluated by the IARC working groups. Other substances from the same sources will be added to the subsequent Annual Reports. The Table of Contents of the Third An nual Report lists all the substances and technological processes included. The list is divided into two sublists. One contains substances, groups of sub stances. or technological processes known to be carcinogens (22 entries). The other list of 95 entries includes substances or groups of substances which may reasonably be anticipated to be carcinogenic. Nickel appears in both sublists because the nickel refin ing process is known to be carcino genic whereas elemental nickel and certain nickel compounds may rea sonably be anticipated to be carcin ogens. A substance not listed in this Annual Report may still be a known carcino gen or a reasonably-anticipated car cinogen. More substances than those included may potentially present a carcinogenic risk to persons living in the United States. These substances will be incorporated into subsequent - Annual Reports as.resources permit. Some chemicals will be introduced as a result of testing performed by in dustry in response to EPA's requests under the Toxic Substances Control ' Act. The NTP will also play a role in providing new information for thescReports as further test results become available. Ionizing radiation, ultraviolet radiatioi (including sunlight), tobacco, alee holic beverages, and some viruses an known or suspected lo be carcinc 9 PCB-ARCH-EXT0378752 gens. They have not been included in this Report because ionizing radiation is discussed thoroughly in a General Accounting Office report, and a recent Surgeon General's report8 reviews the overwhelming relationship between tobacco and cancer~Several publi cations issued by the National Cancer Institute explain the relationship of al coholic beverages, ultraviolet radia tion. and viruses to cancer. varies depending on geographical lo cation and other environmental con ditions. Every year human activities may add more carcinogens to the ambient air. water, food. soil, and occupational environment. Although scientists at tempt to estimate the amount of car cinogens present in and added to the environment, there is no reliable way to predict who is exposed to what car cinogen and when exposure occurs. If exposure is linked to a specific habit Estimating Exposure such as smoking, a specific use such According to clause (ii) of subpara graph (4)(A), this Report is required to include those substances "to which a significant number of people residing in the United States are exposed." Substances to which only a few peo ple are exposed are not included for the most part. Yet some substances which have been banned or restricted in use are contained in the Report (e.g.. safrole, arsenical pesticides. Mirex) as medical diagnosis or treatment, or a restricted environment such as the workplace, our ability to estimate ex posure improves, and in some cases ,i the size of population segments in volved, can be estimated. Production and use data may also help estimates exposure, but this information may not J be publicly available for many sub-* stances, and where it is. it may be out-a| dated or even contradictory. ffi either because people who were pre viously exposed remain potentially at risk, or because these substances are still present in the environment. The National Occupational Hazardii Survey (NOHS). conducted by CDC/Ji NIOSH from 1972 to 1974, gave in-. formation on the number of workers,*! Subparagraph (4)(B) requires that the exposed and on some broader asjjjjffl Annual Reports provide "information pects of occupational exposure. Thffiffiij concerning the nature of such expo forthcoming National Occupational sure and the estimated number of per Exposure Survey (a repeat of NOH^tra sons exposed to such substance." The should yield valuable information witn8|| determination of the number of people respect to more recent exposures3g| exposed and the route, intensity, and Subsequent Annual Reports on CarjglK duration of such exposure for each cinogens will explore this and othe l|j| substance remains a formidable task. available data bases, and theK|||| This Report attempts to respond to these questions, and wherever ade quate answers could be obtained, they strengths and limitations. Regulatory Status iliillPtlll are included. Exposure to carcino gens can occur anywhere. Some car cinogens have always been present in the environment and their amount Subparagraph (4)(C) of Section also requires "a statement (i) each substance contained in theM^^ under subparagraph (A) for whichjj^^^ 10 effluent, ambient, or exposure stan dard has been established by a Fed eral agency. . . The Third Annual Report responds to this requirement by appending to the description of each substance a summary of Federal reg ulations as submitted by the partici pating agencies. Some of these standards and regulations have been enacted for reasons other than the carcinogenicity of the substance, for instance, to prevent other adverse health effects or to improve the envi ronmental quality or food quality. Solid or liquid wastes, or wastes discharged into the air may contain carcinogens, yet these may be regulated as toxic substances or hazardous pollutants and not specifically as carcinogens. But if these regulations reduce expo sure to carcinogens, then the carcin ogenic risk posed by such substances will also decrease. Estimating Risk Reduction Clause (ii) in subparagraph (4)(C) re quires a statement identifying "for each effluent, ambient, or exposure stan dard established by a Federal agency with respect to a substance contained in the list under subparagraph (A), the extent to which, on the basis of avail able medical, scientific, or other data, such standard, and the implementa tion o! such standard by the agency, decrease the risk to public health from exposure to the substance. . . This requires quantified information on the amount of protection from cancer the public receives from established Fed eral standards. Estimating the amount of health pro tection is perhaps the most difficult task in preparing the Annual Reports. One reason is that most Federal laws con cerned with reducing carcinogenic risk have been enacted within the last 15 years. Given the long period between the initial exposure to a carcinogen and the onset of disease, it is still too early to evaluate how much Federal stan dards and other regulations have de creased the risk to public health. Another reason is that information on past exposure levels, which could serve as a baseline for estimating future risk reduction, often is not available or ac curate. The risk--the likelihood of developing cancer--depends on the intensity, route, and duration of exposure to a carcinogen. Individuals may respond differently to similar exposures, de pending on host factors such as age. sex, nutritional status, overall health, and inherited characteristics. Only in a few instances, where studies of long term human exposures and cancer in cidence in restricted environments are available, can risk be estimated with confidence. The regulation of asbestos in the work place provides a good example. In 19G9, under the Walsh-Healy Act which then regulated only the firms with Gov ernment contracts, the standard for permissible exposure was 12 fibers of asbestos per cubic centimeter (cc) of air. Under the Occupational Safety and Health Act, this standard was reduced to 5 fibers/cc in 1972 and to 2 libers/ cc in 1976. On April 17, I960, a joint NIOSH-OSHA Working Group roe-' ommended the elimination of all non essential uses of asbestos and tin; im plementation of a public health pro gram to reduce human exixisures. The high risk to workers exposed prior lo PCB-ARCH-EXT0378753 1969 is well known. The actual data on exposure levels in the `40s, '50s, and '60s are fragmentary, although exposure levels were undoubtedly un acceptably high by today's standards. Mesothelioma (primarily an asbestosinduced cancer) and lung and other cancers, the incidence of which is in creased by exposure to asbestos, generally take 20-40 years or more to develop and become apparent. For these reasons, we will not know until early in the next century to what extent the series of exposure reductions in the 70s have reduced the cancer risk to those persons exposed to asbes tos. One possible way to provide quanti tative estimates of risk reduction might be to assume that the carcinogenic risk is directly proportional to expo sure. This approach also supposes that data on past and present exposure levels are available, or that conditions in all workplaces are in compliance with regulations. However, information supporting these assumptions is only rarely obtainable. Nevertheless, it is reasonable and prudent to accept that the reduction of exposure, for any rea son, particularly to substances shown to be carcinogenic in experimental an imals. will decrease the incidence of cancer. This is the basis of current regulatory policies which aim to lower human exposure to cancer-causing substances and thereby improve pub lic health. For a more detailed discus sion of the issues involved, see the previously cited report to Congress on the assessment of technologies for determining cancer risks from the en vironment.2 References 1. National Center for Health Statis tics. Advance Report of Final Mor tality Statistics. 1979, Monthly Vital Statistics. Report 31, No. 6, Sup plement, September 30. 1982. 2. Office of Technology Assessment. Congress of the United States. As sessment of Technologies for De termining Cancer Risks from the Environment. U.S. Government Printing Office, Washington, D.C., 1981, p. 3 and pp: 65-109. 3. Cole, P. Epidemiological Clues to Host Factors in Human Carcino genesis. In: H. Bartsch and B. Arm strong (Eds.). Host Factors in Human Carcinogenesis. IARC Sci entific Publications No. 39. Inter national Agency for Research on Cancer. Lyon. 1982. p. 3. 4. Bartsch, H. and Armstrong, B. (Eds ). Host Factors in Human Car cinogenesis. IARC Scientific Pub lications No. 39. International Agency for Research on Cancer, Lyon, 1982, p. xi. 5. Griesemer, R.A., and Cueto, C., Toward a Classification Scheme for Degrees of Experimental Evidence - for the Carcinogenicity of Chemi cals for Animals. In: R. Montesano, H. Bartsch and L. Tomatis, (Eds ). Molecular and Cellular Aspects of Carcinogen Screening Tests, IARC Scientific Publications No. 27. In ternational Agency for Research on Cancer. Lyon, 1980, pp. 259-81. 6. International Agency for Research on Cancer. Chemicals. Industrial Processes and Industries Associ ated with Cancer in Humans, IARC Monographs on the Evaluation of Carcinogenic Risk of Chemicals to Humans. Supplement 4. IARC, Lyon. 1982. 7. Comptroller General, U.S. General Accounting Office. Problems of As sessing Cancer Risk of Low Level Ionizing Radiation Exposure Re port to the Congress of the United States, EMD-81-1, Washington, DC.. 1981. 8. Office of Smoking and Health. The Health Consequences of Smoking: Cancer, a Report of the Surgeon General, U.S. Department of Health and Human Services, Public Health Service, Rockville. Maryland, 1982. 9. See, for example, Shimkin, M.B. Science and Cancer, NIH Publi cations No. 80-568, Third Revision, National Cancer Institute, National Institutes of Health, U.S. Depart ment of Health and Human Ser vices, Bethesda. Maryland, 1980. 19 PCB-ARCH-EXT0378754 performed with rats were consid ered inadequate.200 209 i. There are case reports and epi demiological studies of lymphomas occurring in patients that received phenytoin; however, no excess of lymphomas was reported in a fol low-up study of epilepsy patients, many of whom received phenytoin along with other anti-epileptic drugs. Three recent papers report one case of . malignant mesenchymoma and two cases of neuroblastoma in chil dren with phenytoin-induced mal formations. An epidemiological study looked at the frequency of use of phenytoin and of phenobarbitone in mothers of children with childhood cancers compared with mothers of normal children. While more moth ers of cancer cases reported a his tory of epilepsy, no differences were seen in the proportion of epileptic mothers who took either phenytoin or phenobarbitone. An excess of . lymphomas was seen in children of epileptic mothers, but the occur rence of brain tumors was not re ported.200 209 Polybrominated biphenyls (PBBs) Polybrominated biphenyls (PBBs) were widely used as flame retardant addi tives in synthetic fibers and in molded ^International Agency lor Research on Can cer IARC Monographs on (he Evaluation ol the Carcinogenic Risk ol Chemicals to Humans. Supplement 1. Lyon. France: IARC, 1979. ''"International Agency tor Research on Can cer. IARC Monographs on the Evaluation ol the Carcinogenic Risk ol Chemicals to Man. Vol. 13. Lyon. France: IARC. 1977, pp. 201-25. plastics. PBBs are solid at room tem perature and practically insoluble in water. They are mixtures of com pounds containing, on the average, six bromine atoms per molecule. PBBs have been incorporated into the plastic housing of many commer cial products. PBBs have been sold to manufacturers of polycarbonates, polyesters, polyolefins, and polysty renes. The CPSC reported that hexa- bromobiphenyl (HBB) was the major component in the most widely used PBB mixture, and that PBBs are not presently being used in consumer products. The commercial production of PBB began in 1970. Approximately 11 mil lion lb of HBB were manufactured by a single U.S. company from 1970 to 1974 and marketed under the trade names Firemaster BP-6 and Firemas- ter FF-1. CPSC reported that the sole U.S. manufacturer of HBB ceased production in 1974. The remaining stock of this product held by the sole producer was reportedly exhausted in April 1975. . Other PBB production ceased in the late 1970s. EPA reported one U.S. producer (Toxic Substances Control Act (TSCA), Chemical Substance In ventory, 1979, public record). The U.S. production of other PBBs (decabro- mobiphenyl), as reported to the U.S. International Trade Commission, was greater than 10,000 lb for 1980, ac cording to NCI. Workers of a chemical company that manufactured PBB were directly ex posed to PBB. by skin contact, inhal ation, and unintentional ingestion. HBB was detected in workers' sera at a range of 1.1 to 1729 ppb, and in their adipose tissue at a range of 0.51-581 r' 4$ m ;vfL 'M mff. 110 ppm. Approximately 2000 lb of PBBs . were inadvertently mixed into live stock feed in Michigan in 1973. Thou sands of animals died or were destroyed. Initially, 8000 to 12,500 Michigan residents were exposed in 1973 and 1974 to meat, milk, and eggs contaminated with PBBs. A general population survey conducted in Mich igan revealed that only 10 percent of the population did not have detecta ble levels of PBB in their blood. Be cause the PBBs are biologically stable ponent--administered by gavage produced neoplastic nodules and hepatocellular carcinomas in fe male Sherman strain rats.2'0 In an other bioassay, Firemaster FF-1 -- also administered by gavage--was carcinogenic to Fischer 344 rats and B6C3F1 mice of each sex, inducing neoplastic nodules, hepatocellular carcinomas, and cholangiocarcinomas in rats and hepatocellular carcinomas in mice 2" and slowly eliminated, significant body burdens could persist throughout their lifetimes. In 1974, the National Occupational Polychlorinated biphenyls (PCBs) Hazard Survey estimated that 4900 workers were potentially exposed to PBBs. The sources of PBBs currently appear to be residues remaining in and around plants that at one time manu factured, processed, or produced products using PBBs. FDA stated that a potential for human exposure exists from the ingestion of foods containing residues of these compounds. EPA is currently proposing a re porting rule for PBBs under the TSCA, 8(a). Under section 406 of the Food, Drug, and Cosmetic (FD&C) Act. FDA regulates PBB compounds as un avoidable environmental contami nants. In cooperation with CDC and Polychlorinated biphenyls (PCBs) vary in appearance from mobile, oily liq uids to white crystalline solids to hard noncrystalline resins, with all forms in soluble in water. Before 1972, PCBs were used in transformer cooling liq uids, heat transfer and hydraulic fluids, vacuum pump fluids, lubricants, plas ticizers, fillers in investment casting waxes, surface coatings and sealants, pesticide extenders, and for copy pa per. By 1974, all domestic uses of PCBs were confined to closed systems. PCBs have been produced in the United States since 1929. Domestic production reached a peak volume of the Public Health Department of the State of Michigan. FDA monitors the long-term human health effects from acute exposure to PBBs. *,0Kimbrough. R.D.. D.F. Groce. M.P. Korver. and V.W. Burse. Induction ol Liver Tumors in Firemaster FF-1 (Firemaster BP-6 containing 2 percent of calcium tris ilicate)--a mixture of pentabromobiphenyl, hexabromobiphenyl and heptabromobiphenyl with hexabro mobiphenyl being the major com Female Sherman Strain Rats by Polybrominated Biphenyls. J. Nat. Cancer Inst. Vol. 66. No. 3. 1981, pp. 535-42. 211 National Toxicology Program. Studies ol Car cinogenesis Polybrominated Biphenyl Mixture (Firemaster FF-1). Technical Report Series No. 244. NIH Publieniion No. 83-1800. Research Tri angle Park, North Carolina. 1903 86 million lb in 1970 and decreased to approximately 41 million lb by 1974. Annual domestic production prior to 1977 was approximately 35 million lb. Except for limited research and de velopment applications, the PCBs are no longer produced in the United States, and no import or export of the compounds has been permitted since July, 1979. The release of PCBs from prior in dustrial uses and the persistence of the compounds in the environment have resulted in widespread contam ination of water and soil, with subse quent exposure of the general public. Routes of exposure include skin ab sorption and ingestion; a major source of exposure is through the diet. A 1976 report indicated that fish, cheese, eggs, and contaminated animal feed were the major U.S. commodities in which PCBs were found. The percentage of food samples contaminated with PCBs decreased between 1973 and 1975. Residues of PCBs have been de tected in human milk and fat samples collected from the general U.S. pop ulation. EPA reports that approxi mately 12 million persons within 12 miles of three existing and nine proj ected commercial incinerators could be exposed to releases of PCBs. In 1970, EPA banned PCBs as a pesticide ingredient and canceled registration of products containing the PCBs. EPA promulgated toxic pollu tant effluent standards in 1977 and modified the marking and disposal rule in 1979. This latter regulation has been effective in identifying and eliminating some sources. In 1979, EPA issued a ban on manufacturing, processing, and distributing of PCBs in commerce; most manufacturing had ceased voluntarily before these rules were issued. EPA proposed a water quality criteria doc ument that will protect freshwater, salt water. and aquatic organisms, as well as human health. Air pollution sources of PCBs were recently assessed by EPA, and the need for regulation of emissions from incinerators, the only source for which controls may be es sential, is under consideration. Under the Safe Drinking Water Act, EPA is preparing a report to the Congress concerning the nature and extent of PCB contamination, as well as rec ommendations for further regulatory action if necessary. In 1973, FDA established toler ances for PCBs in several foods and in feeds for food-producing animals. In 1979, four of those tolerances were reduced. The current tolerances are 1.5 ppm (fat basis) in milk and manu factured dairy products; 3 ppm (fat basis) in poultry; 0.3 ppm in eggs; 0.2 ppm in finished animal feed; 2 ppm in animal feed components of animal ori gin; 5 ppm in fish and shellfish (edible portion); and 0.2 ppm in infant and junior foods. FDA is currently consider ing the lowering of PCB standards in fish and shellfish from 5 ppm to 2 ppm and has established action levels of 3 ppm (fat basis) in red meat and 10 ppm in paper food-packaging mate rial. OSHA has adopted an 8-hour timeweighted average standard of 1 mg/ m3 for PCBs containing 42 percent chlorine, and 0.5 mg/m3 for PCBs con taining 54 percent chlorine. This stan dard was adopted by OSHA for toxic effects other than cancer. The evidence for carcinogenicity of PCBs in experimental animals is sufficient, and the evidence in hu mans is inadequate.212 Certain mixtures of PCBs (e g., Kanechlor 500, Aroclor 1254 and 1260) were carcinogenic in mice and rats after administration in the diet, producing liver tumors.213 A slight increase in the incidence of cancer, particularly melanoma of the skin, has been reported in a small group of men exposed occupation ally to Aroclor 1254, a mixture of PCBs.212 Procarbazine and procarbazine hydrochloride Procarbazine, an organic compound, is soluble but unstable in water or aqueous solutions. Procarbazine hydrochloride is pri marily used as an antineoplastic agent in the treatment of advanced Hodg kin's disease and oat-cell carcinoma of the lung. The FDA approved the use of procarbazine hydrochloride in 1969 and indicated that the drug should be used as an adjunct to standard ther apy. Possible exposure occurs during manufacture of the drug and direct ex- posure occurs during administration to patients. The National Prescription Audit reported 1.5 million prescrip tions dispensed in 1980. Some of the metabolites of procarbazine are both carcinostatic and carcinogenic. There is sufficient evidence for car cinogenicity in humans of intensive chemotherapeutic regimens that in clude alkylating agents, vinca al kaloids, procarbazine hydrochloride, and prednisone. There is inade quate evidence for carcinogenicity of procarbazine hydrochloride alone in humans. There is sufficient evi dence for the carcinogenicity of procarbazine in experimental ani mals.214215 Laboratory exposure of animals to procarbazine was studied by intraperitoneal injection. In rats, ma lignant lymphoma, adenocarcinoma of the mammary gland, and olfac tory neuroblastomas were induced in statistically significant numbers. In mice, malignant lymphoma or leukemia, olfactory neuroblastoma, alveolar/bronchiolar adenoma, and adenocarcinoma of the uterus were induced in statistically significant numbers.215 Evidence of carcinogenicity was also found in mice and rats following administration by gavage; in rats following intravenous administra tion, and in one instance following transplacental exposure. Two stud ies in two species of nonhuman pri mates suggest that procarbazine hydrochloride may also produce '''International Agency tor Research on Can cer. IARC Monographs on the Evaluation ol the Carcinogenic Risk ot Chemicals to Humans. Supplement 4. Lyon, France: IARC, 1982 '''International Agency for Research on Can cer. IARC Monographs on the Evaluation ol the Carcinogenic Risk ol Chemicals to Humans. Vol 18. Lyon. France: IARC, 1978, pp 43-103 ''"International Agency lor Research on Can cer. IARC Monographs on the Evaluation ol the Carcinogonic Risk ol Chemicals to Humans. Vol 26 Lyon, France IARC, 1981, pp 311-39. '''National Cancer Institute Bioassay ol Pro carbazine (or Possible Carcinoger city Tech nical Report Series No. 19. DHHS Publication No (NIH) 79-819 tiethesda, Maryland, 1979. Executive Summary The Third Annual Report on Carcino gens, prepared by the National Toxi cology Program (NTP), U.S. Public Health Service, is issued by the Sec retary of the Department of Health and Human Services (DHHS) pursuant to Public Law 95-622 of November 9, 1978.* This law requires the Secretary to publish an Annual Report which contains "a list of all substances (i) which either are known to be carcin ogens or which may reasonably be anticipated to be carcinogens and (ii) to which a significant number of per sons residing in the United States are exposed." Annual Reports should also provide available information on the nature of exposures, the estimated number of persons exposed, and the extent to which the implementation of Federal regulations decreases the risk to public health ftom exposure to these substances. For the purpose of this Report, "known carcinogens" are defined as those substances for which the evidence from human studies indicates that there is a causal relationship between expo sure to the substance and human can cer. Substances "which may reasonably be anticipated to be car cinogens" are defined as those for which there is limited evidence ol car cinogenicity in humans or sufficient evidence of carcinogenicity in exper imental animals. Substances in the above categories for which exposure of persons resid ing in the United States has been dem onstrated are included in the Report. Also included are some substances which have been banned, or the use of which has been restricted, because people previously exposed are poten tially at risk, or because some of these substances still remain in the environ ment. The fact that a substance is not contained in the Third Annual Report does not mean that it is not a known or reasonably-anticipated carcino gen. Several problems are encountered in developing the information required for the Annual Reports on Carcinogens. Estimating the number of people ex posed to a substance and identifying the nature, route, and intensity of ex posure are difficult tasks. It is often impossible to obtain accurate produc tion volumes and use patterns for most chemicals. The information that is being collected by the U.S. Environmental Protection Agency under the Toxic Substances Control Act, at least that part which is public record, may assist in obtaining more accurate exposure estimates in the future. In addition, the forthcoming National Occupational Exposure Survey" will provide now in formation on the extent and degree of exposure for different substances and manufacturing -ocesses. Perhaps the hardest task in preparing this Report is to evaluate the effec tiveness of Federal regulations in de creasing the risk to public health. One reason is that virtually all Federal laws concerned with reducing risk have been, enacted within the last 15 years 'An ujviJJite o( the National Occupational tr ard Survey conducted from 1!)7? In I0/-1 h. " r" r v Oi-u v- c v-ifiV " ' PCB-ARCH-EXT0378757